JOURNAL OF NANO- AND ELECTRONIC PHYSICS ЖУРНАЛ НАНО- ТА ЕЛЕКТРОННОЇ ФІЗИКИ

Vol. 4 No 1, 01007(7pp) (2012) Том 4 № 1, 01007(7cc) (2012)

2077-6772/2012/4(1)01007(7) 01007-1 2012 Sumy State University

The Role of Nano-sized Fraction on Spark Plasma Sintering the

Pre-Alloyed Spark-Erosion Powders

G.E. Monastyrsky1*, P. Ochin2, A.V. Gilchuk1, V.I. Kolomytsev3, Yu.N. Koval3

1 NTUU “KPI”, 37, Peremogy Av., UA-03506 Kyiv, Ukraine

2 ICMPE-CNRS, 2-8, rue Henri Dunant-94320 Thiais, France 3 IMP of NASU, 36, Vernadsky Av., UA-03142, Kyiv, Ukraine

(Received 18 October 2011; revised manuscript received 22 February 2012; published online 14 March 2012)

Ti-Ni-Hf, Ni-Al and Cu-Al-Ni shape memory materials were produced by spark plasma sintering

method from the micron and nano-sized particles prepared by spark-erosion method in cryogenic liquid

from preliminary melted master alloys. The effects of spark plasma sintering processing parameters on the

martensitic transformation and microstructure of the sintered compacts were investigated using XRD and

SEM methods. Although precipitating processes were usually not completely depressed, the intensive

grain growth was also not found in most cases. Most of the microstructure peculiarities of as processed

powder were inherited by the sintered material. The contradictory role of the nano-sized fraction of

powders is discussed: in most case this fraction promotes the rapid sintering but also the oxidation

proceses in sintered compacts.

Keywords: Shape memory alloys, Spark-erosion powders, Spark plasma sintering, Ti-Ni, Ni-Al, Cu-Al-Ni.

PACS numbers: 81.05.Bx, 81.20.Ev, 81.30.Kf, 81.30.Mh

______________ * monag@imp.kiev.ua

1. INTRODUCTION

Attractive perspectives of the practical application

the materials with conventional and magnetic shape

memory, high-temperature shape memory stimulate

continuous searching new production and processing

methods. The direction of searching is caused by their

typical characteristics and requirements that are im-

posed to them. In most cases shape memory materials,

high temperature shape memory materials have pre-

cise composition, some of them contain or highly reac-

tive or high-cost components and moreover are a multi-

component. Thus the requirement to their compositions

is especially important. Most of these materials tend to

the decomposition during the heat treatment. In addi-

tion, they, with few exceptions, have poor ductility and

machinability.

In this communication we will focus on three mate-

rials, which are representative most of above men-

tioned peculiarities of functional materials: Cu-Al-Ni-

Ti-Cr, Ni-Al and Ti-Ni-Hf shape memory alloys (SMA).

These kinds of SMA are being developed as ones of the

alternatives for the intermediate and high temperature

applications depending on the alloying content.

Cu-Al-Ni is considered as good candidate for the in-

termediate temperature applications (100-300 C). In

Cu-Al-Ni, during the quenching the metastable phase

undergoes several types of martensitic transformation

(MT) depending on Al content. The characteristics of

the martensitic transformation MT are very sensitive

to the order degree of the phase and the precipitation

processes. Primary 1 precipitation limits the high tem-

perature applications of these alloys modifying the MT.

The eutectic decomposition takes place above 440 C

and finally the + 1 eutectoid decomposition ap-

pears at 500 C. Increasing Al content promotes the

formation 1 phase that embrittles the alloy. In addi-

tion Cu-Al-Ni alloys are brittle due to their very high

elastic anisotropy (A∼13) and large grain size and in

general show poor mechanical properties. Powder met-

allurgy is to be a good method in order to obtain finer

grain size in Cu-Al-Ni SMA.

Enriched with Ni intermetallic Ni-Al is a promising

shape memory alloy (SMA) for high-temperature appli-

cations with the start temperature of martensitic

transformation between 200 and 1200 K depending on

Ni content. Currently, its wide application is limited by

low plasticity, in particular due to the formation of in-

termetallic compounds Ni3Al and Ni5Al3 during the

preparation or heat treatment. By means of the tradi-

tional methods of the production (induction or arc melt-

ing) or by the sintering of elemental powders the ob-

taining of homogeneous alloys, free of these phases, is

not achievable.

Ti-Ni-Hf(Zr) alloys could perform reversible mar-

tensitic transformation at 100-150 C higher than the

room temperature depending on Hf content. Although

the decomposition processes modifying their properties

were not found in the working interval temperature,

the limited ductility, poor machinability and formabil-

ity restrict theirs wide practical applications. As well

as for Cu-Al-Ni and Ni-Al, the powder metallurgy (PM)

is thought to be good alternative method for their pro-

duction and allowing skip over the stages of the shap-

ing parts. Usually the PM production of Ni-Ti-based

alloys from elemental components is accompanying

with the precipitating of extra phases such as Ti2Ni,

Ni4Ti3, Ni3Ti [1], and oxide Ti4Ni2Ox [2]. Therefore the

using of pre-alloyed powders looks as preferable way

for PM processing.

It is the reason why the pre-alloyed spark erosion

(SE) powders of Ti-Ni-Hf, Ni-Al and Cu-Al-Ni-Ti-Cr

obtained in liquid argon were used [3, 4, 5]. Spark ero-

sion is probably the most versatile technique available

for the producing powders of metals, alloys, compounds

with the diameters of particles from a few nm to sever-

G.E. MONASTYRSKY, P. OCHIN, A.V. GILCHUK, ET. AL… J. NANO- ELECTRON. PHYS. 4, 01007 (2012)

01007-2

al dozen µm. Using liquid argon as working liquid, one

can obtain the powders from pre-alloyed material with

a given composition, which is vitally important for

functional material like SMA, free from large quanti-

ties of oxides.

The operation with pre-alloyed powders has unam-

biguous feedback caused by the peculiarities of the

mechanism of powder sintering. The combustion sinter-

ing passing through the eutectic reactions or consump-

tion the heat of formation of intermetallic compounds,

promoting the sintering of elemental powders, is not

available anymore. The mechanism of solid state diffu-

sion sintering between the different powders particles

of the same composition presumes the long term age-

ing. The precipitation processes are superimposed on

the sintering processes during the annealing causing a

decomposition of alloys with the formation of extra

phases. Therefore the spark plasma sintering (SPS)

method looks as a promising alternative of the conven-

tional sintering method of pre-alloyed powders.

SPS method is express method allowing the sinter-

ing powders within several minutes in vacuum or inert

gas atmosphere. It employs simultaneously the rela-

tively high pressure, high temperature of sintering and

heavy current passing through the samples that de-

stroys the oxidation film on the powder surface provid-

ing good sintering. The method is very informative. It

gives “on line” the information about the progress of

the compacting, temperature interval of degassing and

softening of the materials.

The goal of this work was to preliminary study the

applicability of SPS method for the rapid sintering of

pre-alloyed SE powders of Ti-Ni-Hf, Ni-Al and Cu-Al-

Ni-Ti-Cr shape memory alloys.

2. EXPERIMENTAL PROCEDURES

2.1 Spark erosion powder preparation

Commercially pure Ti, Ni, Cu and Hf and Zr (99.9%)

were used for the alloys production by the induction

melting. The rods of Ni49.87-Ti40.25- Hf9.44-Zr0.30-

Cu0.14at.% and Cu-13.01Al-3.91Ni-0.37Ti-0.24Crwt.%

with a diameter of 6 mm was produced by AMT (Bel-

gium). Alloy with a nominal composition Ni63Al37 arc

melted and molded into bars with a diameter of 4 mm

and the length of 12 cm. Part of the rods were used as

electrodes for SE apparatus and the remaining were

broken at 3-4 mm pieces (granules) and were used to

obtain powder by spark erosion method.

The general principle of the spark-erosion pro-

cessing was described in details in [3, 6] and involves

the application of a heavy current between two elec-

trodes and a lot of pieces (granules) that prepared from

the pre-alloyed material both being immersed in liquid

argon inside a container (Fig. 1). The key idea of the

method is the melting (evaporation) of the material by

the electric discharge with a duration of 5-100 sec

with subsequent quenching of molten droplets in liquid

argon in situ. After spark erosion treatment of 100 g of

granules of master alloys about 40 70 g of powder was

received depending on the material composition. In

order to prevent eventual explosion of finest particles of

powders they were kept in a vessel with liquid argon

during the day until all argon was evaporated followed

by were passivated in hexane. Then the powders were

sieved and divided into three fractions: the size of less

than 128 and more than 63 m, less than 63 and more

than 32 m and less than 32 m, designated hereinaf-

ter as 63/125, 32/63 and <32 respectively.

Fig. 1 – Principal scheme of the spark erosion process illus-

trating the formation mechanisms of micron sized and nano-

sized particles during spark erosion processing

Morphology studies of spark erosion powders of all

compositions have shown that micron sized particles

have usually round shape (Fig. 2a). Nano-sized parti-

cles form the agglomerates, which cover the surface of

micron sized particles (Fig. 2a, b).

a

b

Fig. 2 – Morphology of micron sized (a) and nano- sized (b)

particles of spark erosion powders

2.2 Spark plasma sintering processing

Fig. 3 shows the main components of the SPS appa-

ratus (DR.SINTER® LAB Series, Metal Processing

Systems, Inc. Japan). A 50 kN uniaxial press provides

THE ROLE OF NANO-SIZED FRACTION ON SPARK PLASMA SINTERING… J. NANO- ELECTRON. PHYS. 4, 01007 (2012)

01007-3

a mechanical load through the die plungers for densifi-

cation. Simultaneously an AC power supply applies an

alternating current through the sample placed in the

graphite die. About 1 g of powder was preliminary

slightly compacted in the die. A graphite paper was put

between the sample and the graphite plungers. Then

the die was placed inside the working chamber of the

Fig. 3 – A general scheme of spark plasma sintering appa-

ratus

apparatus and the system was evacuated. To avoid

overheating of the sample the heating rate was no

more 350°C/min. Maximum applied uniaxial pressure

was 99.5 MPa. Temperatures of the samples during the

sintering were measured by a sheathed thermocouple,

which was inserted into a small hole in one side of the

graphite mold, as illustrated on Fig. 3. The tempera-

ture, vacuum in the chamber, current applied force,

shrinkage displacements of the powder compacts, volt-

age and current were recorded during the processing.

The electric current provides a rapid Joule heating

in the compact die-sample system. The ramping exper-

iment is initiated with the heating by the application of

continuously increasing electric current together with

the continuously increasing pressure. When the tem-

perature get to preliminary setting temperature of sin-

tering (TSINT) a set constant value of the electric cur-

rent was applied for a given time, hereafter indicated

as sintering time (tS). After that, the sample was un-

loaded, cooled and removed from the die. Temperatures

of sintering were chosen according to the assessed data

of the decomposition, oxidation and others processes

carrying out in material. A typical program of the sin-

tering is shown on Fig. 4. The performed programs of

sintering are listed in Table 1. As a result of sintering a

sintered compacts with a diameter of 8 mm and height

of 4-6 mm were obtained.

Table 1 – The regimes of SPS treatment; F – applied force,

HR – heating rate, d – shrinking displacement. MT means

that sharp inflection of the shrinkage displacement curve (martensitic transformation) has been observed upon the cooling

2.3 Methods of the specimens characterization

XRD study of the compacts were carried out at room

temperature by the Debye-Sherrer method with a

CoK 1,2 radiation. Instrument Philips PW1830 with

Multi-Purpose XRD Diffraction System from PANalyti-

cal was used.

The morphology and microstructure of the sintered

samples were analyzed by LEO 1530 instrument

equipped by PGT PRISM 2000 (Ge) spectrometer with-

out the references standards. Detail composition study

were carried out with JSM-6490LV(Jeol) and JAMP-

9500F (Jeol) both equipped with EDX spectrometer

INCA PentaFETx3 (Oxford Instruments).

3. RESULTS AND DISCUSSION

3.1 SPS process

Fig. 5 illustrates the shrinkage displacement of the

powder compacts and gas pressure in the chamber dur-

ing processing. For all kinds of powders the first stage of

the sample shrinking was accompanied with the dra-

matic vacuum fall obviously related with the plungers’

Regime TSINT,

oC

tS,

min

F,

kN

HR,

oC/min

d,

mm Remark

Cu-Al-Ni

CAN-1000-1 1000 1

5

330 1.60 Melted, 32/63

CAN-700-1 700 1 280 1.97 MT, 32/63

CAN-650-1 650 1 160 1.72 MT, 32/63

CAN-600-1 600 1 150 1.72 MT, 32/63

CAN-480-1 480 1 96 1.52 32/63

CAN-440-1 440 1 88 1.07 32/63

CAN-390-1 390 1 65 1.05 MT?, 32/63

Ni-Al

NA-400-10 400 10

5

200 0.22 32/63

NA-500-10 500 10 250 0.24 32/63

NA-850-10 850 10 212 0.87 MT, 32/63

NA-850-4s 850 4.5 212 0.86 MT, 32/63

NA-900-2 900 2 225 0.92 MT, 32/63

NA-850-4b 850 4.5 212 1.40 MT, 65/125

NA-1000-1H 1000 1 250 1.53 MT, 65/125

Ti-Ni-Hf

TNH-500-0 530 0 2.5

100 0.29 <32

TNH-1000-2 980 2 100 1.20 <32

TNH-860-1H 860 1 5

250 1.19 MT, 32/63

TNH-1000-1H 1000 1 200 1.42 MT, 32/63

Fig. 4 – Typical program of sintering (TSINT 1000 C;

tS=1 min)

G.E. MONASTYRSKY, P. OCHIN, A.V. GILCHUK, ET. AL… J. NANO- ELECTRON. PHYS. 4, 01007 (2012)

01007-4

0

5

10

15

20

25

30

35

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 200 400 600 800 1000

Vacu

um

, P

a

Dis

pla

cem

en

t, m

m

Temperature, oC

Displacement, vaccuum vs time, Ni-Al

d, mm

Vacuum, Pa

Ms ?

Ts

15

17

19

21

23

25

27

29

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 200 400 600 800 1000

Vacu

um

, P

a

Dis

pla

cem

en

t, m

m

Temperature, oC

Displacement, vacuum vs T, Ti-Ni-Hf

d, mm

Vacuum, Pa

Ms ?

Ts

Fig. 5 – Shrinkage displacement of the specimens and gas pres-

sure in the chamber vs. temperature. Programs NA-1000-1,

CAN-700-1 and TNH-1000-1H respectively

motion. At the same time second shrinkage dis-

placement starting at higher temperature seems to be

true sintering of the particles. The temperature of the

first (T1) and second (TS) shrinkage displacement upon

the heating depends on the kind of powder. In case of

Ti-Ni-Hf powder there was third interval too.

Table 2 – Temperature intervals of shrinkage displacement

Material T1 start T1finish TS start TS finish T3 start T3 finish oC

Cu-Al-Ni 120 250 300 >700

Ni-Al 50 400 600-650 >1000

Ti-Ni-Hf 300 425 475 850 910 >1000

Temperature intervals of sintering are listed in Ta-

ble 2.

First sintering interval contributes about 15 %, 30-

40 % and 8-15 % of the total displacement for Cu-Al-Ni,

Ni-Al and Ti-Ni-Hf powders respectively. The value of

contribution depends on the sizes of powders particles.

The more the diameters of particle sizes, the more this

contribution. It is evident that the effect of preliminary

compacting (before the processing) is the less, the more

the particles size. Thus one can assume that this first

interval seems to be connected with the mechanical

compressing and re-arrangement of powder mainly and

only partial welding. It accompanied by the emission of

gas (air), which is kept between the particles and

squeezed out by plungers’ motion. Other processes also

contribute the gas emission. It could be hexane and/or

water desorption (vacuum peak at about 200 C on Ti-

Ni-Hf chart on Fig. 4), collapse of porous in particles and

the chemical reactions realizing the excessive gas(es).

Some of the possible reactions will be discussed below.

It is seen also from Table 1 that the more the diam-

eters of particle sizes (denoted in field “Remark” of Ta-

ble 1 like 32/63 – min and max sizes of particles in

powder fraction), the more total shrinkage of the com-

pact. Thus the second and third sintering interval

doesn’t practically depend on the particles size. The

diffusion controlled processes, promoting high tempera-

ture plasticity of the materials, such as precipitation of

extra phases, decomposition, disordering contribute the

shrinkage displacement effect during the successive

heating.

It is interesting also that upon the cooling the extra

shrinkage appeared, which was displaying as sharp in-

flection on the shrinkage displacement curve (Ms on

Fig. 4, MT in Table 1). Because the position of sharp in-

flection correlate with the martensitic start point of the

respective powder, on can assume that this extra compact-

ing relates with the martensitic transformation stimulat-

ed by the external uniaxial pressure of the plungers.

Mostly powders, even sintered at relatively low

temperatures were mechanically stable. Although spe-

cial mechanical test were not carried out the compacts

were etched and polished like bulk material. Only ex-

clusion was Ni-Al compacts sintered according the pro-

gram NA-400-10 and NA-500-10. It is easy seen from

Table 2 that temperature of sintering (400 and 500oC)

were below second sintering interval, thus confirms the

assumption that after first interval the specimens were

mechanically compacted only. Contrary Cu-Al-Ni pow-

der sintered at 390oC (well above TS and start tempera-

ture of precipitation) was fairly good. However the

best quality of the samples was found for the compacts

sintered at higher temperatures within the interval

650-1000 C depending on the composition of powder.

3.2 SEM investigation

SEM investigation confirmed that after the first

stage of sintering only initial welding and partial com-

pacting appeared. The particles remained mainly non-

deformed with the typical round shape of SE powders

(Fig. 5a). Contrary, the particles after the second stage

of sintering were gathered tight and have been deformed

(Fig. 5b). EDS microanalysis has shown it is not possible

Dis

pla

cem

ent, m

m

Vacu

um

, P

a

THE ROLE OF NANO-SIZED FRACTION ON SPARK PLASMA SINTERING… J. NANO- ELECTRON. PHYS. 4, 01007 (2012)

01007-5

avoid the decomposition processes during SPS pro-

cessing even if the very short time of sintering was used.

Typical for each of powders under investigation extra

phases were found. The type of precipitates depends on

the temperature of sintering mainly. 1 or -phase or

both of them were found in the compacts of Cu-Al-Ni

after different sintering programs (Fig. 7). The X-ray

mapping analyses made for the compacts sintered ac-

cording to the programs CAN-700-1 and CAN-650-1 re-

vealed the redistribution of the alloying elements

(Fig. 8). Two phases are well recognized inside the parti-

cles: matrix phase and precipitated phase sized of sever-

al microns with the composition close to Cu9Al4. Superfi-

cial and interspaces phases between the particles are

also well recognized. The redistribution of the elements

is caused by two different mechanisms of the powder

formation by spark erosion: micron sized particles com-

ing from liquid phase and nano-sized particles coming

from vapor of species through the condensation [3]. Thus

the phase on the surface of particles (superficial) and the

phase from the interspaces between the particles are

rather the mixture of copper and aluminum oxides.

Enriched with Ni phase (perhaps Ni3Al) was found in

Ni-Al powder after NA-850-4s program (Fig. 9a). It

seems to be inherited from the initial microstructure on

Ni-Al SE powders described elsewhere [5]. The inter-

spaces phase between the particles contains many small

a b

Fig. 7 – SE images of Cu-Al-Ni powder after the compacting

programs: (a) CAN-390-1 (arrows indicate phases) and (b)

CAN-700-1 ((arrows indicate 1 phases)

a b

Fig. 6 – SE images of Cu-Al-Ni powder after the compacting

programs CAN-390-1 (a) and CAN-700-1 (b)

Fig. 8 – X-ray elements mapping of the Cu-Al-Ni compact

sintered according to the program CAN-650-1

a b

Fig. 9 – Ni-Al powder after the program NA-850-4s; arrow

indicates the enriched with Ni precipitates (a); morphology of

interspaces’ phase (presumable Al2O3) (b)

Fig. 10 – X-ray elements mapping of the Ni-Al compact sintered according to the program NA-1000-1H

a b

Fig. 11 – Band (martensite) structure in recrystalized Ni-Al

particles program NA-1000-1H (a); band structure resembling

martensite in the Cu-Al-Ni particle (program CAN-700-1) (b)

G.E. MONASTYRSKY, P. OCHIN, A.V. GILCHUK, ET. AL… J. NANO- ELECTRON. PHYS. 4, 01007 (2012)

01007-6

pores (Fig. 9b) and consists of aluminum oxide (Fig. 10).

After NA-1000-1 program some of the particles were re-

crystallized and banded structure resembling the mar-

tensite was also found (Fig. 11a). It is interesting that

banded structure resembling the martensite was also

found in Cu-Al-Ni compacts after etching by Ar+ ions

beam (Fig. 11b).

In addition to the enriched with Ni phase (seems to

be Ti2Ni), a lot of enriched with Hf and oxygen precipi-

tates (with the composition close to HfO2) were found

into Ti-Ni-Hf powders particles as well as on their

boundaries (Fig. 12). Dramatic redistribution of Ni and

Ti was found in interspaces phase between the microns

sized particles (Fig. 13) while Hf is concentrated in mi-

crons particles.

3.3 XRD investigation

XRD study confirmed the results of SEM investiga-

tion. B2, both martensitic 2H and 18R phases, 1 and -

phase or both of them are the main phases in the Cu-

Al-Ni compacts after sintering while in the powder only

2H and B2 were found. In addition, instead of CuO

oxide in powder some noticeable quantity of Cu2O oxide

was found in the sintered compacts.

Although the sintering time was rather short (1-4

min) it does not prevent the formation of the Ni5Al3

phase below 900 C. Ni3Al phase has formed in the sam-

ples sintered at higher temperatures; the lowest amount

of this phase was found for Ni-Al powders treated ac-

cording to the program NA-1000-1. Some minute peaks

observed in sintered compacts as well as in powder could

be considered as belonging to Al2O3 phase.

In addition to the peaks of B2, Ti2Ni and B19’ (mar-

tensite) phases, which are typical for powders’ and

massive material spectra, a lot of HfO2 peaks were

found in Ti-Ni-Hf compacts.

3.4 Effect of nano- fraction on sintering mecha-

nism

The results obtained shown that the spark plasma

sintering method allows sinter the particles of multi

components alloys during extremely short time. Ac-

cording to Fig. 4 the stay of powders above relatively

“safety” temperature of 300 C (from decompositions

and other diffusion controlled processes point of view)

was no more 400 sec. Moreover Cu-Al-Ni and Ti-Ni-Hf

powders were sintered during 1 min even at 390 and

500 C respectively. These temperatures are too low to

provide solid state diffusion controlled sintering mech-

anism. Fig. 12 allows estimating the role of elements

diffusion in mechanism of sintering – the spatial mi-

gration of Ti, Ni, Hf and O atoms is about 1 2 mkm. At

first sight it is more than enough to make available

good adhesion between the particles and thus sintering

mechanism. However clearly observed boundaries be-

tween the particles even in well sintered compacts

(Figures 7b, 11a, 12) refute this statement.

It is considered that applying electrical current dur-

ing the spark plasma sintering produces a cleansing

plasma effect on the surface of the particles [7-10], elec-

tron wind effect (electromigration) [11], an increase in

point defect [12], a decrease in the activation of migra-

tion of the defects. All of these effects can lead to sin-

tering enhancement [13]. In case of spark erosion pow-

der the effect of nano-sized fraction of powders should

be taken into account. The nano-fraction of powders,

filling the interspaces between the micron sized parti-

cles, can enhance the sintering through the highly exo-

thermic reactions between the different species in

nano- fraction producing cohesive, coupling skeleton

from the different type of oxides and alloy.

One of the possible reactions in Ti-Ni-Zr and Ti-Ni-

Hf nano- fraction of spark erosion powders was pro-

posed in [14, 15]. It is highly exothermic redox reaction

between NiO and TiO oxides: TiO + NiO TiO2 + Ni,

which appears in nano powders. As it is seen from

Fig. 12, 13 very reactive Hf can modify the path of re-

action in case of micron sized particles however with

Fig. 12 – Ti-Ni-Hf powder after the compacting programs

TNH-1000-2. Bright spots and needles are the enriched with Hf and oxygen phase

Fig. 13 – X-ray elements mapping of the Ti-Ni-Hf powder sintered according to the program TNH-1000-2

THE ROLE OF NANO-SIZED FRACTION ON SPARK PLASMA SINTERING… J. NANO- ELECTRON. PHYS. 4, 01007 (2012)

01007-7

the same result – the formation of Hf and Ti oxides.

Also the redox reaction can take place in Cu-Al-Ni

nano- powders as was discussed in [16]. In that case

two possible type of reaction are possible:

1) – 4CuO + 2Al + 1/2O 2Cu2O + Al2O3 and

2) – 2Al + 6CuO = 3Cu2O + Al2O3. Both of reactions

are highly exothermic resulting in the mixture of ox-

ides binding the micron sized particles. It is not clear

the possible mechanism (if any) of oxide formation in

Ni-Al powder because of traces of Al2O3 oxide was

found already in powder. However this oxide is a domi-

nant phase filling the interspaces between the micron

sized particles as well produced from nano- fraction

and one can expect that similar (may be more complex)

reaction is responsible for the sintering mechanism.

As a result of above mentioned reactions complex

composite structure is formed from the oxides and oth-

er products with embedded micron sized particles.

Thus the role of the nano- spark erosion powders seems

to be double: from one hand the presence of nano- pow-

ders promotes the sintering, from the other one it is

hard to expect good shape memory characteristics from

such complex composite (alloy + oxides).

4. CONCLUSIONS

The compacts of Ni49.87-Ti40.25-Hf9.44-Zr0.3-

Cu0.14at.%, Cu-13.01Al-3.91Ni-0.37Ti-0.24Crwt.% and

Ni63Al37 shape memory alloys were elaborated by the

spark plasma sintering method from the micron, sub-

micron and nano- sized particles prepared by spark-

erosion method in liquid argon from preliminary melt-

ed master alloys.

Although the strongly shrinkage displacement has

been observed already at relatively low temperature

(< 300-400 C), true sintering has appeared in the com-

pacts sintered at higher temperatures.

Rapid sintering prevents the intensive grain growth

in most cases and provides the inheritance of most mi-

crostructure peculiarities of as-received powder by the

sintered material. Even the sintering time was rather

short (1-10 min) it does not prevent the precipitation of

extra phases.

Highly exothermic reactions between nano- frac-

tions (NiO/TiO/Hf), (CuO/Al) can promote from one

hand the sintering, from other hand the oxidation pro-

cesses during the SPS sintering.

The martensitic transformation in most of the sin-

tered compacts has been clearly indicated during the

cooling in spark-plasma apparatus; martensitic struc-

ture was confirmed by SEM and XRD study.

AKNOWLEDGEMENTS

The authors are grateful to CNRS PICS-3717,

STCU-3520 and ECONET projects for the supporting of

this work.

REFERENCES

1. G.E. Monastyrsky, J. Van Humbeeck, V.I. Kolomytsev,

Y.N. Koval, Intermetallics 10, 613 (2002).

2. M. Nishida, C.M. Wayman, T. Honma, Metall. Trans. A

17, 1505 (1986).

3. G.E. Monastyrsky, P.A. Yakovenko, V.I. Kolomytsev,

Yu.N. Koval, A.A. Shcherba, R. Portier, Mater. Sci. Eng. A

481-482, 643 (2008).

4. G.E. Monastyrsky, P. Ochin, G.Y. Wang, A.V. Gilchuk,

V.I. Kolomytsev, Yu.N. Koval, V.O. Tinkov, A.A. Shcherba,

S.M. Zaharchenko, Chem. Met. Alloys 4, 126 (2011).

5. G.E. Monastyrsky, V.V. Odnosum, V.І. Kolomitsev,

Yu.N. Koval, P. Ochіn, R. Portier, A.A. Shcherba,

S.N. Zaharchenko, Metallofiz. Novejsh. Tehnol. 30, 761

(2008).

6. A.E. Berkowitz, M.F. Hansen, F.T. Parker, K.S. Vecchio,

F.E. Spada, E.J. Lavernia, R. Rodriguez, J. Magn. Magn.

Mater. 254-255, 1 (2003).

7. M. Tokita, Mater. Sci. Forum 308-311, 83 (1999).

8. Z. Shen, M. Johnsson, Z. Zhao, M. Nygren, J. Amer. Ce-

ram. Soc. 85, 1921 (2002).

9. M. Omori, Mater. Sci. Eng. A 287, 183 (2000).

10. D. Zhang, Z. Fu, R. Yuan and J. Guo, Multiphased Ceram-

ic Materials (Ed. W.H. Tuan and J.K. Guo) (Springer: Ber-

lin: 2004).

11. H.B. Huntington, Diffusion in Solids (Ed. A.S. Nowick and

J.J. Burton (Academic Press: New York: 1975).

12. P. Asoka-Kumar, M. Alatalo, V.J. Gosh, A.C. Kruseman,

B. Nielson, K.G. Lynn, Phys. Rev. Lett. 77, 2097 (1996).

13. Z.A. Munir, U. Anselmi-Tamburini, J. Mater. Sci. 41, 763

(2006).

14. O.M. Іvanova, M.І. Danilenko, G.E. Monastyrsky,

V.І. Kolomytsev, Yu.N. Koval, A.A. Scsherba,

S.M. Zaharchenko, R. Portier, Metallofiz. Novejsh. Tehnol.

31, 603 (2009).

15. G.E. Monastyrsky, V.І. Kolomytsev, Yu.N. Koval,

P. Ochin, G. Vang, O. M. Ivanova, N.I. Danilenko,

Metallofiz. Novejsh. Tehnol. 33, 289 (2011).

16. P. Ochin, A.Yu. Pasko, G.E. Monastyrsky, A.V. Gilchuk,

V.I. Kolomytsev, Yu.N. Koval, R. Portier, Spark plasma

sintering of Cu-Al-Ni shape memory alloy, to be published

in J. Alloy Compd.